StuffNet: Using 'Stuff' to Improve Object Detection
This work addresses object detection challenges for computer vision researchers, offering incremental improvements through the integration of segmentation features.
The paper tackles the problem of improving object detection, particularly for small objects, by incorporating features from 'stuff' segmentation, resulting in a performance increase from 18.8% to 23.9% mAP on Pascal VOC 2010.
We propose a Convolutional Neural Network (CNN) based algorithm - StuffNet - for object detection. In addition to the standard convolutional features trained for region proposal and object detection [31], StuffNet uses convolutional features trained for segmentation of objects and 'stuff' (amorphous categories such as ground and water). Through experiments on Pascal VOC 2010, we show the importance of features learnt from stuff segmentation for improving object detection performance. StuffNet improves performance from 18.8% mAP to 23.9% mAP for small objects. We also devise a method to train StuffNet on datasets that do not have stuff segmentation labels. Through experiments on Pascal VOC 2007 and 2012, we demonstrate the effectiveness of this method and show that StuffNet also significantly improves object detection performance on such datasets.